Standardized outputs

Konflux is using standardized outputs in various places, this part of docs will provide more detail information about it. Each one of the results described below is rendered in the UI, you simply click on task name and panel with details is showed containing all results from the task.

TEST_OUTPUT

Each Tekton task in a build PipelineRun provides a TEST_OUTPUT result (JSON) that summarizes its outcome. It is then evaluated by the Conforma team to determine whether the product can be released.

This result type can also be used in integration test pipelines and tasks where it can be parsed by the integration service when deciding the outcome of the integration test. This means you can use it to fail individual test tasks by emitting this result instead of failing the task and stopping the entire pipeline. With this approach it is possible to run multiple tests within the same pipeline and pass or fail them individually.

The result field in TEST_OUTPUT supports the following values: SUCCESS, FAILURE, ERROR, and WARNING. If at least one task produces a WARNING outcome and no tasks produce FAILURE or ERROR, the overall integration test outcome is WARNING. This outcome is treated as passed, but the integration service reports it with a "passed with warnings" message to git providers.

The note field is an optional string that provides additional context about the test outcome. When present, the note is displayed in the Note column of the test report table posted to git provider pull or merge requests.

SCAN_OUTPUT

This one is located within build definition task clair scan. SCAN_OUTPUT result[json] is displayed in different manner than TEST_OUTPUT, but it serves the same purpose. It additionaly displays found vulnerabilities, patched(RHSA fix) and unpatched ones. Should be used in addition to TEST_OUTPUT.

Vulnerabilities
Scan results within Konflux will only report vulnerabilities associated with RHSAs as fixable. This can lead to inconsistencies with Clair scan results found in Quay or elsewhere.

IMAGES_PROCESSED

Result[json] for recording which image digests are processed (i.e. multi-arch images). This enables the ability to validate that all of the image manifest architectures have been scanned even if a task does not support scanning all image manifests.